trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 19.876712328767123,
  "eval_steps": 10.0,
  "global_step": 80,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2191780821917808,
      "grad_norm": 0.2932123008987269,
      "learning_rate": 1.25e-05,
      "loss": 0.008783930912613869,
      "memory(GiB)": 17.37,
      "step": 1,
      "token_acc": 0.995092693565976,
      "train_speed(iter/s)": 0.019999
    },
    {
      "epoch": 2.4383561643835616,
      "grad_norm": 2.0700027278327098,
      "learning_rate": 4.923500664848326e-05,
      "loss": 0.013703253534105089,
      "memory(GiB)": 21.34,
      "step": 10,
      "token_acc": 0.997708053426065,
      "train_speed(iter/s)": 0.024416
    },
    {
      "epoch": 4.876712328767123,
      "grad_norm": 0.6635928259983686,
      "learning_rate": 4.4728512734909844e-05,
      "loss": 0.01251780241727829,
      "memory(GiB)": 21.34,
      "step": 20,
      "token_acc": 0.9980889867475385,
      "train_speed(iter/s)": 0.024506
    },
    {
      "epoch": 7.438356164383562,
      "grad_norm": 0.3758404879079538,
      "learning_rate": 3.689868482592684e-05,
      "loss": 0.009205681085586549,
      "memory(GiB)": 21.34,
      "step": 30,
      "token_acc": 0.9988561809392266,
      "train_speed(iter/s)": 0.024403
    },
    {
      "epoch": 9.876712328767123,
      "grad_norm": 0.16592426655101397,
      "learning_rate": 2.7064483636808313e-05,
      "loss": 0.006611303985118866,
      "memory(GiB)": 21.34,
      "step": 40,
      "token_acc": 0.9993527333531109,
      "train_speed(iter/s)": 0.024512
    },
    {
      "epoch": 12.438356164383562,
      "grad_norm": 0.20857252507816335,
      "learning_rate": 1.6882513269882917e-05,
      "loss": 0.0060650348663330075,
      "memory(GiB)": 21.34,
      "step": 50,
      "token_acc": 0.9996614988829463,
      "train_speed(iter/s)": 0.024404
    },
    {
      "epoch": 14.876712328767123,
      "grad_norm": 0.3868665089464845,
      "learning_rate": 8.067960709356478e-06,
      "loss": 0.004569733142852783,
      "memory(GiB)": 21.34,
      "step": 60,
      "token_acc": 0.9996116859681583,
      "train_speed(iter/s)": 0.024495
    },
    {
      "epoch": 17.438356164383563,
      "grad_norm": 0.09910480250109534,
      "learning_rate": 2.1056668336235622e-06,
      "loss": 0.004885964840650558,
      "memory(GiB)": 21.34,
      "step": 70,
      "token_acc": 0.9996539792387543,
      "train_speed(iter/s)": 0.024481
    },
    {
      "epoch": 19.876712328767123,
      "grad_norm": 0.06850686355051422,
      "learning_rate": 0.0,
      "loss": 0.004139205440878868,
      "memory(GiB)": 21.34,
      "step": 80,
      "token_acc": 0.9997290142371751,
      "train_speed(iter/s)": 0.024486
    }
  ],
  "logging_steps": 10,
  "max_steps": 80,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.42154125149995e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}